Sex-related communication with mothers and close friends during the transition to university.

When people imagine college students talking about sex and sex-related topics, they are likely to picture them talking to their friends. However, although a great deal is known about college students' sex-related communication with their mothers, much less is known about such communication with friends. In the current study, we examine sex-related communication with mothers and with close friends, with the specific goals of understanding (a) differences in the frequency and quality of communication with mothers compared with close friends, (b) gender differences in sex-related communication, (c) how communication with mothers and close friends relates to sexual experience and beliefs. This study therefore provides both descriptive information about the experiences of sex-related communication for new college students and a better understanding of how this communication is associated with behaviors and beliefs.

Christopher (2001) examined parent and peer socialization of sexuality from a symbolic interactionist perspective, arguing that as individuals reach late adolescence, parental influence lessens and peers become more important. He also argues (as we have elsewhere; see Lefkowitz & Gillen, 2005) that late adolescent/emerging adult sexuality includes not only behaviors but also attitudes and beliefs. Christopher argued that sexual beliefs at this stage are multidimensional and complex and include dimensions such as conservative/ permissive attitudes about sex and beliefs about instrumentality. On the basis of this perspective, we predict that students in their 1st year at university will have more frequent and better quality communication with close friends than with mothers and that their sexual experiences and beliefs will be more closely associated with communication with friends than with mothers. In this study, we focus on communication in the past 3 months because we are interested in recent contextual experiences. Clearly, earlier experiences of communication with parents during childhood and adolescence are likely to be associated with sexual behaviors and beliefs in the college years (e.g., Lehr, DiIorio, Dudley, & Lipana, 2000). However, in the current article we are specifically interested in recent experiences of communication and their association with behaviors and beliefs.

Sex-Related Communication With Mothers and Friends

Research suggests that the most frequently discussed topics between mothers and adolescents or emerging adults are sexual dangers and safety, dating and sexual behavior, pregnancy, abstinence, and menstruation (girls only; DiIorio, Kelley, & Hockenberry-Eaton, 1999; Feldman & Rosenthal, 2000; Pistella & Bonati, 1998). Least commonly discussed with mothers are wet dreams, solitary sexual activity, and birth control (boys only; DiIorio et al., 1999; Feldman & Rosenthal, 2000). Similar to conversations with parents, topics discussed most frequently with friends are dating, behaviors and feelings, and reproductive health (DiIorio et al., 1999; Lefkowitz, Boone, & Shearer, 2004; Pistella & Bonati, 1998). Topics less frequently discussed with friends are HIV, rape, and abstinence. (Lefkowitz et al., 2004).

High quality sex-related communication has been conceptualized as conversations in which both partners feel relatively open, comfortable, and are not avoidant or overly embarrassed/negative in discussing these topics (DiIorio et al., 1999; Feldman & Rosenthal, 2000; Lefkowitz et al., 2004). Research suggests that quality of communication is generally better with mothers than with fathers and that adolescents are more comfortable discussing sex-related topics with friends than with parents (DiIorio et al., 1999; Feldman & Rosenthal, 2000). Adolescents and emerging adults feel relatively comfortable and open discussing sex-related topics with friends (DiIorio et al., 1999; Kallen, Stephenson, & Dougherty, 1983; Lefkowitz et al., 2004).

Fewer studies have compared adolescents' and emerging adults' communication with mothers versus with friends or peers. As suggested by Christopher (2001), findings reveal that both male and female youth discuss sex-related topics with their friends more than with their parents (Handelsman, Cabral, & Weisfeld, 1987; Papini, Farmer, Clark, & Snell, 1988; Rozema, 1986), although during mid-adolescence, sex-related communication with mothers may be more common than with friends (DiIorio et al., 1999). What remains unanswered, however, is whether these differences in frequency are specific to certain topics or whether they exist across all sex-related communication. That is, differences between mothers and friends may be due to the topics most frequently discussed with friends, such as intercourse and dating, but may not generalize to other topics, such as abstinence or sexual dangers. Understanding these partner differences may help to better describe the recent experiences of 1st-year college students.

Gender Differences in Sex-Related Communication

Girls and young women generally report more frequent sex-related communication with their parents than do boys and young men (Feldman & Rosenthal, 2000; Fisher, 1987; Raffaelli, Bogenschneider, & Flood, 1998). It remains unclear whether these gender differences are topic specific. There are also known gender differences in the quality of sex-related communication with parents. Girls tend to report feeling more comfortable and open discussing these topics with mothers than do boys (DiIorio et al., 1999; Kotchick, Dorsey, Miller, & Forehand, 1999).

Similar, although not universal, gender differences exist for sex-related communication with peers or friends. DiIorio et al. (1999) found that 13-15-year-old boys talked to their friends about sex-related topics less than did girls, whereas Papini et al. (1988) did not find gender differences for 14-18 year olds. Although gender differences in specific topics are infrequently examined, there is some evidence that female emerging adults discuss a range of topics with their close friends more than do male emerging adults (Lefkowitz et al., 2004).

Associations Between Sex-Related Communication and Sexual Behavior

Some researchers have found links between adolescent sexual experience and frequency (Pick & Andrade Palos, 1995) or quality (East, 1996; Handelsman et al., 1987; Lehr et al., 2000; Miller, Norton, Fan, & Christopherson, 1998) of sex-related communication with parents, such as an association between more frequent conversations with mothers about birth control and a greater likelihood of adolescents being sexually active (Jaccard, Dittus, & Gordon, 1996). However, Whitaker, Miller, May, and Martin (1999) reported that more extensive communication with parents about sex-related topics was associated with a greater likelihood of condom use. Others have failed to find associations between frequency of communication and sexual risk behaviors (Kotchick et al., 1999). Thus, the direction of this association depends on the participants' age and gender and the measures of communication and behavior. As with gender differences, it is unclear whether these differences are universal across all sex-related topics or whether they are specific to certain topics of discussion.

Research examining sex-related communication with peers or friends and sexual experience has yielded more consistent findings. More frequent communication about sex-related topics with close friends is associated with a higher likelihood of being sexually active (DiIorio et al., 1999; Lefkowitz et al., 2004; Treboux & Busch-Rossnagel, 1990). Evidence does suggest, however, that certain topics, such as dating, may not differ on the basis of sexual experience and that others, such as abstinence, may be more common for sexually abstinent than for sexually active individuals (Lefkowitz et al., 2004). Thus, findings for associations between communication with friends and sexual experience are more consistent than are findings for communication with mothers. However, because these associations have rarely been examined in the same study and because of differences in measures used and participants' age, it is unclear whether these differences in communication with friends versus with mothers will emerge if examined within the same study.

Associations Between Sex-Related Communication and Sexual Attitudes

Sexual and condom attitudes are important predictors of individuals' sexual behaviors (Boone & Lefkowitz, 2004; McLaughlin, Chen, Greenberger, & Biermeier, 1997; Pleck, Sonenstein, & Ku, 1993; Yep, 1993). In addition to general attitudes about sexuality, attitudes about condoms are important because of their links with condom use. In particular, attitudes about condoms' outcomes and self-efficacy for condom use or beliefs about preparedness and capability for using condoms have been related to actual condom use (Baele, Dusseldorp, & Maes, 2001; Basen-Engquist et al., 1999). Thus, understanding attitudes about sex and condoms can provide insight into individuals' future sexual behavior, particularly for individuals who are not currently sexually active. Attitudes may be particularly important at the transition to university, when individuals are living in a new environment and are exposed to new peers and opportunities. Sexual attitudes are known to be changing around this time, on average becoming more liberal and less guilt-oriented (Chara & Kuennen, 1994; Lefkowitz, 2005; Miller & Moore, 1990). The changing nature of attitudes in this period makes it likely that these topics are discussed with others and that such conversations may influence current and future attitudes.

Some researchers have found that youth who discuss sex-related topics with their parents endorse abstinence more or have more conservative attitudes; others have found that those who discuss sex-related topics more have more positive attitudes about contraception (Miller et al., 1998; Papini et al., 1988). Still others report that for older adolescents communication with mothers relates to similarity of mother-adolescent sexual attitudes (Fisher, 1986). In one comparative study, communication with peers was associated with more liberal attitudes, but communication with parents was associated with more conservative attitudes (DiIorio et al., 1999). More frequent discussion with friends of topics such as fertility issues, behaviors and feelings, dangers, and contraception is associated with less conservative attitudes about sex and more positive and self-efficacious attitudes about condoms (Lefkowitz et al., 2004). Thus, the function of sex-related communication between individuals and their mothers compared with that between individuals and friends may differ. The finding that communication with mothers is associated with more conservative attitudes than is communication with friends (DiIorio et al., 1999) suggests that communication with mothers serves a restricting or protective function in that mothers are likely to encourage abstinence and conservatism, whereas communication with friends may serve an endorsing function in that conversations with peers may normalize or encourage sexual freedom. However, because past research has not examined differences in specific topics that individuals discuss with mothers versus with friends, it is unclear the extent to which these different messages are communicated.

In summary, on the basis of the theoretical perspective of Christopher (2001) and past findings, our specific hypotheses were that:

1. Students would report more frequent and better quality communication about sex-related topics with their close friends than with their mothers.

2. Women would report more frequent and better quality sex-related communication with their mothers and their close friends than would men.

3. Sexually active students would report more frequent and better quality communication than would abstinent students, with the exception of frequency of discussing abstinence.

4. More frequent and better quality communication (with the exception of abstinence discussions) would be associated with more liberal attitudes about sexuality and more positive and self-efficacious attitudes about condoms. However, it was predicted that these associations would be stronger for communication with close friends than with mothers and that, for some measures, communication with mothers could be associated with less liberal or positive attitudes.

Method

Participants

In August 2002, the Registrar's office of a large, public university provided information on all students who began their 1st year that summer or fall. All students age 17-19 who were Black or Hispanic (Registrar's definition), and a random 9% of all students age 17-19 who were Caucasian (Registrar's definition), were contacted about participating in the study. From these initial 839 students, 52% agreed to participate. Surveys were administered in group sessions. Consent was obtained at the beginning of the session, and participants received $25. The total sample size for this larger study was 434. After participating in the first session, a random subsample of participants was invited to participate in a second session. Only students who were at least 18 by the date of the second session were included. During this session, participants engaged in videotaped tasks in pairs (videotaped data not used in current analyses) and filled out surveys at the end of the session. These participants received an additional $25. At the first survey, participants had been at university for an average of 48.1 days (SD = 27.2). The number of days between the two visits averaged 23.6 (SD = 11.4).

The current sample consists of the 182 individuals (50% female) who participated in the second session. On the basis of both the Registrar's records and self-report, 25% of the sample were Black (including African American, African, Caribbean), 23% were Latino (including Mexican American, Puerto Rican, and South American), and 50% were White (not Latino). Participants ranged in age from 17 to 19 (M = 18.5, SD = 0.4) at their first visit. The majority lived in the dorms with roommates (98%). For 69% of the sample, participants' mothers were married to their fathers.

Participants who identified as gay, lesbian, or homosexual were excluded from the video phase of the study for reasons related to the videotaped tasks. Therefore, the vast majority of participants identified as heterosexual (99%), with less than 1% identifying as bisexual.

Measures

Sex-related communication. At the first visit, participants completed a 37-item questionnaire assessing frequency and quality of sex-related communication with their closest same-sex friend in the past 3 months. To identify this friend, women were asked "Please think of the one female at [University] whom you consider to be your closest friend. If you have more than one female close friend at [University], please just pick ONE of these friends." Men were asked the same question for a male close friend. They reported on this friend's age and the length of their friendship. Close friends were an average of 18.2 years old (range = 17 to 24, SD = 0.9). About half (49%) of participants reported having been friends for 3 months or less. At the second visit, participants were asked the same 38 questions about their mothers.

The frequency measure included 24 items that were rated on a 4-point scale (range: never, once, a few times, often), with higher scores indicating more frequent communication. Twenty frequency items had been used in an earlier study with a sample from the same university (Lefkowitz et al., 2004). An additional 4 items that did not address sexual issues (homework, classes, your family, entertainment [e.g., movies, TV shows]) were added in order to compare sex-related communication with communication about everyday topics.

We used the same scales used in earlier work (Lefkowitz et al., 2004): four subscales and three individual items (condoms, contraception, and abstinence). The subscales were dating (four items; e.g., "dating/romantic relationships), fertility issues (four items; e.g., "pregnancy"), behaviors and feelings (five items; e.g., "sexual desire"), and dangers (four items; e.g., "sexually transmitted diseases"). Reliabilities were similar to those in the original study (men: [alpha] =.63 to .86 friend, .75 to .92 mother; women: [alpha] =.67 to .83 friend, .69 to .85 mother). The four new items were considered to be communication about everyday issues. These items had somewhat lower reliability (men: [alpha] = .63 friend; .84 mother; women: [alpha] = .54 friend; .58 mother). Correlations among frequency subscales ranged from. 13 to .64 for friends and from .06 to .74 for mothers.

The measure of quality of communication about sex-related topics was adapted from a measure of communication quality with parents (Feldman & Rosenthal, 2000) and was used previously in its current version (Lefkowitz et al., 2004). The 13 items addressed comfort, openness, and lack of embarrassment when discussing sex-related topics. Feldman and Rosenthal (2000) demonstrated that these constructs are associated with adolescents' evaluation of their mothers as sex educators. Participants rated each item on a 4-point scale (range: strongly disagree to strongly agree), with higher scores indicating better quality. Earlier work suggested that all items about close friends loaded well as one factor (Lefkowitz et al., 2004), and reliability in the current sample was comparable (men: [alpha] = .88 friend; .90 mother; women: [alpha] = .88 friend; .91 mother).

Sexual experience. Participants answered the following question: "Have you ever had penetrative sex (sex in which the penis penetrates the vagina or anus)?" Participants who responded "yes" were considered sexually active, and those who responded "no" were considered sexually abstinent.

Sexual attitudes. A 12-item version of the Sexual Attitudes Scale (Hudson, Murphy, & Nurius, 1983) was used to assess participants' general conservative attitudes toward sex (e.g., "I think sex should be reserved for marriage" and "There is too much sex on television"). Pilot testing indicated that the 12-item and longer versions were highly correlated (r = .96). Respondents rated their agreement on a 5-point scale (range: strongly disagree to strongly agree), with higher scores indicating more conservative attitudes. Reliability on the short measure in the current sample ([alpha] = .88 men, .86 women) was similar to that reported in previous research ([alpha] = .94; Hudson et al., 1983).

Attitudes about condoms. The Outcome Expectancies of Condom Use Scale (Jemmott & Jemmott, 1992) was used to measure participants' positive attitudes about condoms. The Prevention Expectancies subscale consists of three items, with higher scores indicating stronger beliefs in condoms' ability to provide physical protection from pregnancy and disease (e.g., "Condoms can prevent sexually transmitted diseases"). The Hedonistic Expectancies subscale consists of five items, with higher scores indicating more positive feelings about condoms (e.g., "Sex feels good when a condom is used"). Although Jemmott and Jemmott (1992) used the scale with younger adolescents, we have also found that hedonistic expectancies are associated with college students' traditional attitudes about gender (Shearer, Hosterman, Gillen, & Lefkowitz, 2005) and that both hedonistic and prevention expectancies are associated with sex-related communication with friends (Lefkowitz et al., 2004) in an older college sample. Items were rated on a 5-point scale for prevention expectancies, anchors were strongly disagree to strongly agree; for hedonistic expectancies, anchors were strongly disagree to strongly agree for 4 questions, and very negative to very positive for 1 question. Jemmott and Jemmott report adequate reliability for this scale ([alpha] = .50 - .73). Reliability was also adequate in the sample studied here ([alpha] = .91 for prevention, .72 for hedonistic men; [alpha] = .86 for prevention, .78 for hedonistic women).

Condom use self-efficacy. Participants reported their feelings of self-efficacy regarding condom use by answering nine questions. Two of the subscales were from the Sexual Risk Behavior Beliefs and Self-Efficacy Scales (Basen-Engquist et al., 1999). The Perceived Barriers to Condom Use subscale uses a 4-point scale, with higher scores indicating negative beliefs or attitudes about condom use (e.g., "It would be embarrassing to buy condoms in a store"). The Buying and Using Condoms subscale (e.g., "How sure are you that you could use a condom correctly or explain to your partner how to use a condom correctly?") uses a 5-point scale, with higher scores indicating better self-efficacy. The communication about condom use sub-scale is from the Condom Use Self-Efficacy Scale (Barkley & Burns, 2000), and uses a 4-point scale, with higher scores indicating worse communication self-efficacy (e.g., "...how sure are you that you could tell your partner you want to start using condoms?"). For perceived barriers, anchors were strongly disagree to strongly agree; for buying and using, anchors were not sure at all to totally sure; and for communication, anchors were strongly disagree to strongly agree. Past research has reported adequate reliability for these scales ([alpha] = .66 - .73; Basen-Engquist et al., 1999; Barkley & Burns, 2000). Reliability was similar in the sample studied here ([alpha] = .51 to .81 men; .58 to .81 women).

Results

Differences Between Communication With Mothers Versus With Close Friends

Differences between frequency of communication with mothers versus with close friends were examined in a series of 2 (partner, within-subject) by 2 (gender, between-subjects) mixed-method analyses of variance (ANOVAs). In these ANOVAs, an F value of 1.8 corresponds to a small effect size ([eta] = 0.1), an F value of 17.3 corresponds to a medium effect size ([eta] = 0.3), and an F value of 58.3 corresponds to a large effect size ([eta] = 0.5). The conversational partner effect was significant in six out of eight instances, with mostly medium to large effect sizes (see Table 1). Participants described talking to mothers between once and a few times about dating compared with between a few times and often with friends. For behaviors and feelings, condoms, and contraception, participants described discussing these issues less than once with mothers and once to slightly more than once with friends. Although fertility issues were discussed significantly more with friends than with mothers, the means were quite similar and were less than once for both. Finally, participants described discussing nonsexual issues between a few times and often with both mothers and close friends though significantly more frequently with friends.

A two-way mixed ANOVA was also performed to examine differences in communication quality. Results revealed that participants reported higher quality communication about sex-related topics with close friends than with mothers, F(1, 176) = 272.8, p < .001, a large effect size. On average, participants agreed with statements indicating openness and comfort in sex-related discussions with close friends (M = 3.2, SD = 0.4) and chose responses slightly closer to disagreeing than agreeing for communication with their mothers (M = 2.4, SD = 0.6).

Eight correlations were performed to examine the associations between frequency of communication about each sex-related topic with mother and with close friend. These correlations were all significant and indicated low-to-moderate associations. Thus, individuals who reported discussing a particular topic more frequently with mothers also indicated discussing these topics more frequently with friends. The association was lowest for abstinence (r =. 18, p < .05), and highest for fertility issues (r = .52, p < .001). Quality of communication with mothers and with close friends was also positively associated (r = .27, p < .001).

Gender Differences

The ANOVA performed to examine partner differences also revealed that women discussed five of the eight topics more frequently than did men, with a range from small to large effect sizes (see Table 1). Specifically, women reported discussing dating, fertility issues, contraception, abstinence, and nonsexual topics more frequently than did men. There was a significant interaction between gender and speaker for behaviors and feelings. Follow up t tests indicated that men discussed behaviors and feelings significantly more than did women with close friends, t(175) = 1.97, p < .05, but frequency with mothers did not differ (p < .05).

The two-way ANOVA indicated a significant gender difference in communication quality, F(1,176) = 8.4, p < .01, a small effect size. Women reported feeling slightly more comfortable and open talking to their close friends and mothers about sex-related topics than did men.

Associations Between Communication and Sexual Experience

To examine differences by sexual experience at the bivariate level, eight ANOVAs were performed to compare the 108 sexually active students with the 74 abstinent students (see Table 2). In each ANOVA, conversational partner was a within-subject factor, and sexual experience (sexually active vs. abstinent) and gender were between-subjects factors. Because we were interested in sexual experience differences in these analyses, we report on the main effects of sexual experience and any interactions between sexual experience and the other independent variables. There were significant main effects, with relatively small effect sizes, of sexual experience in all ANOVAs except for the one with nonsexual topics. Sexually active students reported talking more about most topics, but less about abstinence, than did abstinent students. There were no interactions with gender, but there were three significant interactions with conversational partner. Follow-up t tests for fertility, behaviors and feelings, and contraception revealed that there were no significant differences between sexually active and sexually abstinent individuals' communication with their mothers (ts < 1.8, ps > .05), but there were differences in communication with close friends (ts > 3.6, ps < .001). An additional ANOVA was performed for communication quality. There was a significant main effect of sexual experience, F(1, 174) = 6.7, p < .01, as well as an interaction between sexual experience and conversational partner, F(1, 174) = 10.1, p < .01. Follow-up t tests revealed that communication quality with close friends was greater for sexually active (M = 3.4, SD = 0.4) than for abstinent individuals (M = 3.0, SD = 0.4), t = 5.2, p < .001, but did not differ for communication with mothers (t< 1).

Two logistic regressions were performed to determine how sex-related communication with close friends (see Table 3) and mothers (see Table 4) differentiated sexually active versus abstinent students. For both regressions, gender (coded 0 for men, 1 for women) was entered in the first step, and the eight sex-related communication measures were entered in the second step. In both instances, communication was a significant predictor of sexual experience. The model for close friend communication correctly classified 77% of the sample, and for mother communication it correctly classified 67%. In the friend model, students who discussed behaviors and feelings more frequently were more likely to be sexually active than were those who discussed these topics less frequently. In contrast, students who discussed abstinence more frequently with friends were less likely to be sexually active compared with those who discussed abstinence less frequently. Finally, individuals who described more open and comfortable communication with friends were more likely to be sexually active than were those with less open communication.

The significant predictors were slightly different in the model of communication with mothers. Individuals who discussed dating with their mothers more frequently were more likely to be sexually active. As with communication with friends, however, students who discussed abstinence more frequently with their mothers had a lower likelihood of being sexually active.

Associations Between Communication and Sexual and Condom-Related Attitudes

Regressions were performed to examine how the eight measures of communication with close friends (see Table 5) and mothers (see Table 6) were related to attitudes about sex and condoms after controlling for gender and in the context of the other communication measures. The change in [R.sup.2] between the step with gender and the step with the communication variables was significant in seven out of the eight friend regressions and approached significance for the eighth. The communication measures explained between 8% and 30% of the variance in the attitudinal measures. Communication quality was a significant predictor in all eight friend regressions. Individuals who reported more open and comfortable communication with their friends had less traditional attitudes about sex and more positive condom attitudes and self-efficacy.

Frequency of communication with close friends about dating significantly predicted two of the attitudinal measures. Discussing dating more frequently with friends was associated with less positive condom attitudes and self-efficacy. More frequent discussion about fertility issues marginally predicted self-efficacy for using and buying condoms. In contrast to the finding for dating, more frequent communication about fertility with friends was related to greater condom use self-efficacy. More frequent communication about behaviors and feelings with close friends significantly predicted two of the condom measures. Students who talked with their friends more frequently about sexual behaviors and feelings felt more self-efficacious for using and buying condoms, but they more strongly believed that condoms interfere with sexual pleasure. Individuals who discussed sexual dangers more frequently with friends had less conservative attitudes about sex. Individuals who discussed condoms more frequently were marginally less certain of condoms' preventive ability. In contrast, individuals who discussed contraception more frequently were marginally more certain of condoms' preventive ability. Discussion of abstinence with friends was a significant predictor in three of the regressions. Individuals who discussed abstinence more frequently tended to have more conservative sexual attitudes, less confidence in condoms' protective ability, and less self-efficacy for using and buying condoms.

Unlike the regressions with communication with friends as the predictors, adding communication with mothers resulted in significant change in [R.sup.2] in only three of the eight models, explaining between 10% and 19% of the variance. No communication measure was significant in all regressions. Discussing dangers was marginally significantly associated with belief in condoms' preventive ability; individuals who discussed sex-related dangers with their mothers more frequently tended to have more confidence in condoms' preventive abilities. In contrast, individuals who discussed condoms with their mothers more frequently tended to have less confidence in condoms' protective abilities. Individuals who discussed abstinence with their mothers more frequently tended to have more conservative sexual attitudes.

Finally, quality of communication with mothers was associated with hedonistic beliefs about condoms and was marginally associated with perceived barriers to using condoms. Individuals who had more open and comfortable communication with their mothers were less concerned about condoms interfering with sexual pleasure and perceived fewer barriers to condom use.

Discussion

Differences Between Communication With Mothers Versus With Close Friends

Findings revealed that, overwhelmingly, students talked to their close friends more than to their mothers about sex-related topics in the past 3 months. These findings are consistent with work suggesting that adolescents discuss these topics more with friends than with parents (Handelsman et al., 1987; Papini et al., 1988). Although there were also significant differences in discussions of nonsexual topics, the effect sizes were smaller than for dating or for behaviors and feelings. These differences support Christopher's (2001) contention that peers play an important role in sexuality at this age. There were not, however, differences between discussing abstinence and sexual dangers with mothers versus with friends. These topics may be similarly discussed with mothers and friends because they involve the restriction or negative aspects of sexuality. It is interesting, however, that although discussed equally with mothers and with friends, abstinence was one of the most frequent topics with mothers and one of the least frequent topics with friends. Thus, the experience of discussing abstinence has different meanings when the conversational partner is the mother or the close friend, even if they occur at similar rates.

There was also a significant difference in communication quality between close friends and mothers. Participants were more likely to feel comfortable and open during sex-related communication with friends than with mothers, paralleling earlier work with adolescents (DiIorio et al., 1999). Thus, differences between mothers and close friends are not only in the topics discussed but also in how they feel discussing these topics. It may be that more frequent communication leads to more comfort in discussions or that more comfort leads to more frequent discussions. Given the cross-sectional nature of these data, we cannot examine the direction of the association. However, the greater frequency and comfort with friends suggests that friends may play an important part in assigning meaning to sexual roles and behaviors (Christopher, 2001).

Despite these differences, there were also associations across conversational partner. Topics discussed more frequently with mothers were also discussed more frequently with close friends. Even though sex-related communication may be partner specific, there is also individual variation. That is, an individual's experiences and attitudes may be reflected in conversations with multiple partners. The association between quality of communication with mothers and with close friends suggests that some individuals may generally feel more comfortable with sex-related topics, influencing their communication across conversational partners.

Gender Differences

As is frequently reported in the literature (e.g., Feldman & Rosenthal, 2000; Lefkowitz et al., 2004; Raffaelli et al., 1998), there were also gender differences in communication. Women reported discussing several sexual topics more frequently than did men. Women also reported discussing nonsexual topics more frequently than did men, but the effect size was smaller than for dating or fertility issues. The fact that gender differences were consistent in mother and close friend relationships (with the exception of conversations about behaviors and feelings) suggests that gender differences may not be partner specific. That is, women discuss sex-related topics more frequently than do men, whether they are with their mothers or with their friends. Of course, the current study limited the gender of parent (mothers only) and close friend (matched to participant) on the basis of evidence that these would be the relationships with the most extensive discussions of sex-related topics. In future work, it will be important to include fathers, as well as opposite-sex friends, in order to further understand gender differences at the dyadic level.

Associations Between Communication and Sexual Experience and Attitudes

Past work has consistently found associations between sex-related communication with friends and sexual experience and attitudes (DiIorio et al., 1999; Lefkowitz et al., 2004; Treboux & Busch-Rossnagel, 1990) but has been more mixed when examining communication with mothers (Jaccard et al., 1996; Kotchick et al., 1999; B. C. Miller et al., 1998). In the current study, there were a number of associations between communication and sexual experience, particularly for communication with friends. In addition, communication with friends was a stronger correlate of sexual attitudes than was communication with mothers. It is interesting that communication with mothers and with friends had the strongest association with general sexual attitudes and the weakest for self-efficacy for communicating about condoms. We are not sure why the general sexual attitudes measure was strongly linked, whereas self-efficacy for the most related behavior--communicating with one's partner--was associated more weakly by communicating with close others. This measure did have the lowest psychometric properties of all attitudinal measures, so it is possible that it is a measurement problem rather than a true lack of association.

Discussions of abstinence were associated with a decreased likelihood of being sexually active and with more conservative sexual attitudes. Discussing abstinence with close friends was also associated with less positive views about condoms. Thus, individuals who are practicing abstinence and who endorse conservative attitudes are discussing this topic with their friends and with their mothers more than those who are sexually active. Research with younger adolescents suggests that mothers discuss abstinence in the context of other safer sex techniques and HIV (Lefkowitz, Boone, Au, & Sigman, 2003). However, given the higher incidence of discussing abstinence among sexually abstinent compared with sexually active individuals, it is likely that by the transition to university, discussions reflect behavior rather than the reverse. Future longitudinal designs would also help to address this question.

The discussion of behaviors and feelings with close friends, but not with mothers, was associated with an increased likelihood of being sexually active. It may be that discussing one's commitment to abstinence arises in the context of the mother-child relationship, but that discussing one's sexual involvement is less likely to be brought up. It is interesting to note that discussing dating with mothers was associated with an increased likelihood of being sexually active. One possible explanation is that sexually active individuals are more likely to be in romantic relationships. However, dating discussions with friends did not relate to sexual experience, suggesting that for sexually active students, discussing dating with mothers may be a safer topic than is sexual behavior. It is surprising that individuals who discussed dating more frequently with their friends tended to have more negative views about condoms. Because this finding differed from other findings, and was opposite of our predictions, it should be interpreted with caution. One possible new direction to tease apart these complex processes might be to examine how communication with mothers or friends explains the association between self- and partner's attitudes, as has been done for research with mothers in the past (Fisher, 1986).

There were other instances in which associations were not in the direction predicted. For example, individuals who discussed condoms more with close friends were marginally less trusting of condoms' preventive abilities. This finding was opposite of the finding for communication with mothers, which was associated with more trust in condoms' preventive abilities. It is possible that discussions of condoms with mothers are more didactic, with mothers suggesting that condoms are important; whereas discussions with friends could include individuals' doubts and concerns about condoms' safety to prevent pregnancy and/or disease. Given the importance of beliefs about condoms for individuals' condom use behavior (Boone & Lefkowitz, 2004; Lewis, Malow, & Ireland, 1997), these associations should continue to be explored. The direction of association across all of our frequency measures, however, highlights the importance of examining specific sex-related topics. Some of the inconsistent findings in past research may be due to different measurement across studies.

The Importance of Communication Quality

An interesting finding was that the quality of communication with close friends was related to sexual experience and all sexual and condom attitudes. In contrast, quality of communication with mothers was related to only one condom attitude (and marginally for a second). It is possible that comfort and openness in discussing sex-related topics with friends has more importance for sexual attitudes than does comfort and openness with mothers. That is, given that the context of close friendships is more similar to romantic and sexual relationships than is the context of the mother-child relationship, comfort and openness in discussing sex-related topics with friends may translate into comfort and openness with one's sexual relationships more generally. It is also possible, however, that individuals who are more sexually experienced, and/or are more liberal and positive about sexuality and condoms, are more likely to feel comfortable talking about these topics with their friends but that sexual experience and attitudes do not have as much effect on comfort and openness of discussions with mothers. That is, the quality of discussions with mothers might be less dependent on the individual's own experience and attitudes than might be discussions with friends.

There were some limitations to the current study. First, the residential college experience is unique, and we expect that findings would differ for emerging adults who still live at home or who live with a romantic partner. The experience of comfortable sex-related communication with friends may be an aspect of the college experience that does not generalize to other situations. Second, because of study design, all participants completed the questions about their friends before the questions about their mothers. Thus, it is possible that order effects explain the differences over time. However, we would have expected that filling out questions about friends earlier in the semester could lead to lower rates of sex-related communication than filling them out later in the semester, when the participants might know their friends for longer. Third, two of our self-efficacy measures demonstrated less than optimal reliability, which may explain why communication about sex did not explain much variance in self-efficacy for communicating about condoms. Fourth, the current findings are cross-sectional. Longitudinal data are necessary to begin to understand how communication may predict changes in experience and attitudes and how new behaviors (e.g., having sex for the first time) may predict changes in communication. Finally, our measures of communication, although more detailed than in many past studies, still do not tell us much about the process of communication. In future work, it will be interesting to understand more specific messages conveyed about these topics. For instance, discussions about condoms could address either the importance of using condoms or doubts about condoms' effectiveness.

Despite these limitations, the current study revealed important findings about sex-related communication with mothers and with close friends. During the transition to university, students discussed sex-related topics more frequently with close friends than with mothers and felt more comfortable doing so. Women in general discussed these topics more frequently and with more comfort and openness than did men. Sex-related communication, particularly the quality of communication and particularly that with close friends, was associated with both sexual experience and attitudes. These findings suggest the importance of understanding the role of friends in assigning meaning to sexual behaviors and beliefs.

This work was supported by National Institute of Child Health and Human Development Grant #R-01 HD 41720 to Eva S. Lefkowitz. We are grateful to Meghan Gillen, Shelley Hosterman, Eric Loken, Susan McHale, Lisa Meyer, Kristie Patton, Cindy Shearer, Lyndsey Sturm, and Amber Thompson for their help with study design, data collection, data scoring and entering, data cleaning and statistical analyses, and to Annie Rissman and Tara Stoppa for comments on an earlier version of this article.

References

Baele, J., Dusseldorp, E., & Maes, S. (2001). Condom use self-efficacy: Effect on intended and actual condom use in adolescents. Journal of Adolescent Health, 28, 421-431.

Barkley, T. W., & Burns, J. L. (2000). Factor analysis of the Condom Use Self-Efficacy Scale among multicultural college students. Health Education Research, 15, 485-486.

Basen-Engquist, K., Masse, L. C., Coyle, K., Kirby, D., Parcel, G. S., Banspach, S., & Nodora, J. (1999). Validity of scales measuring the psychosocial determinants of HIV/STD-related risk behavior in adolescents. Health Education Research, 14, 25-38.

Boone, T. L., & Lefkowitz, E. S. (2004). Safer sex and the health belief model: Considering the contributions of peer norms and socialization factors. Journal of Psychology and Human Sexuality, 16, 51-68.

Chara, P. J., & Knennen, L. M. (1994). Diverging gender attitudes regarding casual sex: A cross-sectional study. Psychological Reports, 20, 427-440.

Christopher, F. S. (2001) To dance the dance: A symbolic interactional exploration of premarital sexuality. Mahwah, NJ: Lawrence Erlbaum Associates.

DiIorio, C., Kelley, M., & Hockenberry-Eaton, M. (1999). Communication about sexual issues: Mothers, fathers, and friends. Journal of Adolescent Health, 24, 181-189.

East, P. L. (1996). The younger sisters of childbearing adolescents: Their attitudes, expectations, and behaviors. Child Development, 67, 267-282.

Feldman, S. S., & Rosenthal, D. A. (2000). The effect of communication characteristics on family members' perceptions of parents as sex educators. Journal of Research on Adolescence, 10, 119-150.

Fisher, T. D. (1986). An exploratory study of parent-child communication about sex and the attitudes of early, middle, and late adolescents. Journal of Genetic Psychology, 147, 543-557.

Fisher, T. D. (1987). Family communication and the sexual behavior and attitudes of college students. Journal of Youth and Adolescence, 16, 481-495.

Handelsman, C. D., Cabral, R. J., & Weisfeld, G. E. (1987). Sources of information and adolescent sexual knowledge and behavior. Journal of Adolescent Research, 2, 455-463.

Hudson, W. W., Murphy, G. J., & Nurius, P. S. (1983). A short-form scale to measure liberal vs. conservative orientations toward human sexual expression. Journal of Sex Research, 19, 258-272.

Jaccard, J., Dittus, P. J., & Gordon, V. V. (1996). Maternal correlates of adolescent sexual and contraceptive behavior. Family Planning Perspectives, 28, 159-185.

Jemmott, L. S., & Jemmott, J. B. (1992). Increasing condom-use intentions among sexually active Black adolescent women. Nursing Research, 41, 273-279.

Kallen, D. J., Stephenson, J. J., & Dougherty, A. (1983). The need to know: Recalled adolescent sources of sexual and contraceptive information and sexual behavior. Journal of Sex Research, 19, 137-159.

Kotchick, B. A, Dorsey, S., Miller, K. S., & Forehand, R. (1999). Adolescent sexual risk-taking behavior in single-parent ethnic minority families. Journal of Family Psychology, 13, 93-102.

Lefkowitz, E. S. (2005). "Things have gotten better": Developmental changes among emerging adults after the transition to university. Journal of Adolescent Research, 20, 40-63.

Lefkowitz, E. S., Boone, T. L., Au, T. K., & Sigman, M. (2003). No sex or safe sex? Mothers' and adolescents' discussions about abstinence and safer sex. Health Education Research, 18, 341-351.

Lefkowitz, E. S., Boone, T. L., & Shearer, C. L. (2004). Communication with close friends about sex-related topics during emerging adulthood. Journal of Youth and Adolescence, 33, 339-351.

Lefkowitz, E. S., & Gillen, M. M. (2005). "Sex is just a normal part of life": Sexuality in emerging adulthood. In J. J. Arnett & J. L. Tanner (Eds.), Coming of age in the twenty-first century: The lives and contexts of emerging adults (pp. 235-255). Washington, DC: American Psychological Association.

Lehr, S. T., Dilorio, C., Dudley, W. N., & Lipana, J. A. (2000). The relationship between parent-adolescent communication and safer sex behaviors in college students. Journal of Family Nursing, 6, 180-196.

Lewis, J. E., Malow, R. M. & Ireland, S. J. (1997). HIV/AIDS in heterosexual college students: A review of a decade of literature. Journal of American College Health, 45, 147-158.

McLaughlin, C. S., Chen, C., Greenberger, E., & Biermeier, C. (1997). Family, peer, and individual correlates of sexual experience among Caucasian and Asian American late adolescents. Journal of Research on Adolescence, 7, 33-53.

Miller, B. C., & Moore, K. (1990). Adolescent sexual behavior, pregnancy, and parenting: Research through the late 1980s. Journal of Marriage and the Family, 52, 1025-1044.

Miller, B. C., Norton, M. C., Fan, X., & Christopherson, C. R. (1998). Pubertal development, parental communication and sexual values in relation to adolescent sexual behaviors. Journal of Early Adolescence, 18, 27-52.

Papini, D. R., Farmer, F. L., Clark, S., & Snell, W. E. (1988). An evaluation of adolescent patterns of sexual self-disclosure to parents and friends. Journal of Adolescent Research, 3, 387-401.

Pick, S., & Andrade Palos, P. (1995). Impact of the family on the sex lives of adolescents. Adolescence, 30, 667-675.

Pistella, C. L. Y., & Bonati, F. A. (1998). Communication about sexual behavior among adolescent women, their families, and their peers. Families in Society: The Journal of Contemporary Human Services, 79, 206-211.

Pleck, J. H., Sonenstein, F. L., & Ku, L. C. (1993). Masculinity ideology: Its impact on adolescent males' heterosexual relationships. Journal of Social Issues, 49, 11-29.

Raffaelli, M., Bogenschneider, K., & Flood, M. F. (1998). Parent-teen communication about sexual topics. Journal of Family Issues, 19, 315-333.

Rozema, H. J. (1986). Defensive communication climate as a barrier to sex education in the home. Family Relations, 35, 531-537.

Shearer, C. L., Hosterman, S. J., Gillen, M. M., & Lefkowitz, E. S. (2005). Are traditional gender role attitudes associated with risky sexual behavior and condom-related beliefs? Sex Roles, 52, 311-324.

Treboux, D., & Busch-Rossnagel, N. A. (1990). Social network influences on adolescent sexual attitudes and behaviors. Journal of Adolescent Research, 5, 175-189.

Whitaker, D. J., Miller, K. S., May, D. C., & Martin, L. L. (1999). Teenage partners' communication about sexual risk and condom use: The importance of parent-teenager discussions. Family Planning Perspectives, 31, 117-121.

Yep, G. A. (1993). Health beliefs and HIV prevention: Do they predict monogamy and condom use? Journal of Social Behavior & Personality, 8, 507-520.

Eva S. Lefkowitz and Graciela Espinosa-Hernandez

Pennsylvania State University 

Correspondence should be addressed to Eva S. Lefkowitz, Department of Human Development and Family Studies, 110 South Henderson Building, The Pennsylvania State University, University Park, PA 16802. E-mail: [e-mail] 

Table 1. Analyses of Variance of Frequency of Communication by
Conversational Partner and Gender

Item                      Close Friend          Mother

                            M        SD       M        SD
Dating
  Male                     2.0      0.7      1.0      0.9
  Female                   2.5      0.6      1.7      0.7
Fertility Issues
  Male                     0.4      0.7      0.3      0.6
  Female                   1.3      0.8      1.1      0.7
Behaviors and Feelings
  Male                     1.3      0.9      0.3      0.4
  Female                   1.1      0.8      0.3      0.5
Dangers
  Male                     0.6      0.8      0.5      0.8
  Female                   0.7      0.7      0.7      0.7
Condoms
  Male                     1.0      1.1      0.6      1.0
  Female                   1.1      1.0      0.6      1.0
Contraception
  Male                     0.8      1.1      0.4      0.7
  Female                   1.1      1.1      0.9      1.0
Abstinence
  Male                     0.6      1.0      0.6      1.0
  Female                   0.8      1.1      0.9      1.1
Nonsexual topics
  Male                     2.5      0.5      2.2      0.7
  Female                   2.7      0.3      2.5      0.4

Item                      F (Partner)       F (Gender)

Dating                    207.9 ****        45.1 ****
  Male
  Female
Fertility Issues
  Male                      5.4 **         100.0 ****
  Female
Behaviors and Feelings    201.9 ****         0.80
  Male
  Female
Dangers                     0.90             1.20
  Male
  Female
Condoms                    23.0 ****         0.00
  Male
  Female
Contraception              11.2 ****        13.9 ****
  Male
  Female
Abstinence                  0.2              4.0 **
  Male
  Female
Nonsexual topics           26.4 ****        16.2 ****
  Male
  Female

Item                      F (Interaction)

Dating                    1.96
  Male
  Female
Fertility Issues
  Male                    0.6
  Female
Behaviors and Feelings    4.6 **
  Male
  Female
Dangers                   1.1
  Male
  Female
Condoms                   0.1
  Male
  Female
Contraception             1.1
  Male
  Female
Abstinence                0.3
  Male
  Female
Nonsexual topics          0.1
  Male
  Female

Note. Sample size ranges from 177 to 179 because of missing data;
scale range: 0 = never, 3 = often.

** p < .05. **** p < .001.

Table 2. Analyses of Variance of Frequency of Communication by Sexual
Experience and Conversational Partner

Item                      Sexually Active   Sexually Abstinent

                            M        SD        M        SD

Dating                     1.9      0.6       1.6       0.7
Fertility Issues           0.9      0.6       0.6       0.7
  Close friend             1.0      0.3       0.6       0.8
  Mother                   0.8      0.7       0.6       0.8
Behaviors and Feelings     0.9      0.5       0.6       0.5
  Close friend             1.4      0.9       0.9       0.8
  Mother                   0.3      0.5       0.4       0.5
Dangers                    0.7      0.6       0.5       0.6
Condoms                    1.0      0.8       0.6       0.8
Contraception              1.0      0.8       0.6       0.8
  Close friend             1.2      1.1       0.6       0.9
  Mother                   0.8      1.0       0.5       0.9
Abstinence                 0.6      0.7       0.9       0.9
Nonsexual topics           2.5      0.4       2.5       0.4

Item                               F (Sexual Experience)

Dating                             5.4 **
Fertility Issues                   6.8 ***
  Close friend
  Mother
Behaviors and Feelings            15.0 ****
  Close friend
  Mother
Dangers                            4.8 **
Condoms                           13.5 ****
Contraception                      9.9 ***
  Close friend
  Mother
Abstinence                         7.4 ***
Nonsexual topics                   1.0

Item                               F (Interaction
                                   Between Sexual
                                   Experience and
                                   Conversational Partner)

Dating                             1.3
Fertility Issues                   4.2 **
  Close friend
  Mother
Behaviors and Feelings            11.9 ****
  Close friend
  Mother
Dangers                            0.1
Condoms                            0.1
Contraception                      7.6 ***
  Close friend
  Mother
Abstinence                         0.6
Nonsexual topics                   0.5

Note. Means are only broken down by conversational partner when the
interaction between sexual experience and conversational partner was
significant. Sample size ranges from 177 to 179 because of missing
data. Scale range: 0 = never, 3 = often.

** p < .05. *** p < .01. **** p < .001.

Table 3. Logistic Regression Analyses Predicting Sexual Experience
(Sexually Active Versus Sexually Abstinent) From Gender, and
Sex-Related Communication With Close Friends

Step and Variable          B            SE B           Wald
                                                   [chi sqaure]

Step 1
  Gender                 -0.55          .31            3.19 *

Step 2
  Gender                 -0.41          .55            1.61
  Dating                 -0.45          .36            1.61
  Fertility Issues        0.73          .44            2.84
  Behaviors and           0.76          .64            4.49 **
  Feelings
  Dangers               -0.52           .46            1.24
  Condoms                0.29           .28            1.10
  Contraception          0.00           .26            0.00
  Abstinence            -0.78           .22           12.52 ***
  Quality                1.72           .52           10.72 ***

Step and Variable     Odds Ratio       Step
                                    [chi sqaure]

Step 1                                 3.22 *
  Gender                 0.58

Step 2                                55.93 ***
  Gender                 0.66
  Dating                 0.64
  Fertility Issues       2.08
  Behaviors and          2.14
  Feelings
  Dangers                0.60
  Condoms                1.40
  Contraception          1.00
  Abstinence             0.46
  Quality                5.57

Note. N = 179.

* p < .07. ** p < .05. *** p < .001.

Table 4. Logistic Regression Analyses Predicting Sexual Experience
(Sexually Active Versus Sexually Abstinent) From Gender, and
Sex-Related Communication With Mothers

Step and Variable          B            SE B          Wald
                                                   [chi square]

Step 1
  Gender                 -0.60          .31            3.78 **

Step 2
  Gender                 -0.64          .45            2.02
  Dating                  0.64          .26            5.94 **
  Fertility Issues       -0.22          .46            0.23
  Behaviors and          -0.39          .67            0.65
  Feelings
  Dangers                 0.59          .40            2.20
  Condoms                 0.37          .28            1.69
  Contraception          -0.20          .26            0.57
  Abstinence             -0.56          .20            8.14 ***
  Quality                -0.26          .32            0.68

Step and Variable      Odds Ratio      Step
                                    [chi sqaure]

Step 1                                3.82 **
  Gender                  0.55

Step 2                               20.18 ***
  Gender                  0.53
  Dating                  1.89
  Fertility Issues        0.80
  Behaviors and           0.68
  Feelings
  Dangers                 1.81
  Condoms                 1.44
  Contraception           0.82
  Abstinence              0.57
  Quality                 0.77

Note. N = 178.

** p < .05. *** p <.01.

Table 5. Standardized Betas From Regressions Predicting Sexual
Attitudes From Gender and Sex-Related Communication With Close Friends

Step and Variable                Conservative
                                   Sexual
                                  Attitudes

Step 1
  Gender                             .10
Step 2
  Gender                             .20 **
  Dating                            -.08
  Fertility Issues                  -.02
  Behaviors and Feelings            -.03
  Dangers                           -.21 **
  Condoms                            .10
  Contraception                      .01
  Abstinence                         .34 ****
  Quality                           -.46 ****
    Step 1 [R.sup.2]                 .01
    Step 2 [R.sup.2]                 .31 ****
    [DELTA][R.sup.2](Step 2)         .30 ****

Step and Variable                        Condom Outcomes

                                  Prevention       Hedonistic

Step 1
  Gender                            -.08               .12
Step 2
  Gender                            -.04               .03
  Dating                            -.21 **            .12
  Fertility Issues                  -.09              -.17
  Behaviors and Feelings             .02              -.27 *
  Dangers                            .11              -.04
  Condoms                           -.21 *             .17
  Contraception                      .20 *             .12
  Abstinence                        -.17 **           -.02
  Quality                            .34 ****          .35 ***
    Step 1 [R.sup.2]                 .01               .02
    Step 2 [R.sup.2]                 .16 ****          .14 **
    [DELTA][R.sup.2](Step 2)         .15 ****          .13 **

Step and Variable                   Condom Use Self-Efficacy

                                  Use & Buy      Communication

Step 1
  Gender                            -.26 ***          -.21 **
Step 2
  Gender                            -.34 ****         -.14
  Dating                            -.17 **            .00
  Fertility Issues                   .22 *             .05
  Behaviors and Feelings             .24 **            .17
  Dangers                           -.14              -.12
  Condoms                            .02              -.05
  Contraception                      .07               .09
  Abstinence                        -.15 **           -.01
  Quality                            .30 ****         -.32 ****
    Step 1 [R.sup.2]                 .07 ****          .04 ***
    Step 2 [R.sup.2]                 .31 ****          .12 ***
    [DELTA][R.sup.2](Step 2)         .24 ****          .08 *

Step and Variable               Condom Use Self-Efficacy

                                   Barriers

Step 1
  Gender                             .01
Step 2
  Gender                             .05
  Dating                             .10
  Fertility Issues                  -.06
  Behaviors and Feelings            -.19
  Dangers                            .18
  Condoms                           -.17
  Contraception                     -.02
  Abstinence                         .05
  Quality                           -.35 ****
    Step 1 [R.sup.2]                 .00
    Step 2 [R.sup.2]                 .22 ****
    [DELTA][R.sup.2](Step 2)         .22 ****

Note. Sample size ranges from 171 to 179 because of missing data.

* p < .07. ** p < .05. *** p < .01. **** p < .001.

Table 6. Standardized Betas from Regressions Predicting Sexual
Attitudes From Gender, and Sex-Related Communication With Mothers

Step and Variable               Conservative
                                  Sexual
                                 Attitudes

Step 1
  Gender                            .10
Step 2
  Gender                            .11
  Dating                            .00
  Fertility Issues                 -.02
  Behaviors and                    -.03
    Feelings
  Dangers                          -.07
  Condoms                           .00
  Contraception                    -.15
  Abstinence                        .42 ****
  Quality                          -.07
    Step 1 [R.sup.2]                .01
    Step 2 [R.sup.2]                .20 ****
    [DELTA][R.sup.2] (Step 2)       .19 ****

Step and Variable                    Condom Outcomes

                                 Prevention      Hedonistic

Step 1
  Gender                            -.04            .16 *
Step 2
  Gender                            -.01            .12
  Dating                            -.03            .13
  Fertility Issues                  -.10           -.07
  Behaviors and                     -.02           -.12
    Feelings
  Dangers                            .25 *          .12
  Condoms                           -.30 **        -.05
  Contraception                      .01            .08
  Abstinence                        -.10           -.15
  Quality                            .03            .18 **
    Step 1 [R.sup.2]                 .00            .03 *
    Step 2 [R.sup.2]                 .06            .09
    [DELTA][R.sup.2] (Step 2)        .06            .06

Step and Variable               Condom Use Self-Efficacy

                                 Use & Buy     Communication

Step 1
  Gender                           -.24**          -.23 **
Step 2
  Gender                           -.24**          -.29 ***
  Dating                            .12            -.06
  Fertility Issues                 -.10             .15
  Behaviors and                     .00             .12
    Feelings
  Dangers                           .02            -.19
  Condoms                           .19            -.09
  Contraception                     .09             .08
  Abstinence                       -.14             .00
  Quality                           .09            -.13
    Step 1 [R.sup.2]                .06 **          .05 **
    Step 2 [R.sup.2]                .15 ****        .10 **
    [DELTA][R.sup.2] (Step 2)       .10 **          .05

Step and Variable               Condom Use Self-Efficacy

                                  Barriers

Step 1
  Gender                            .00
Step 2
  Gender                            .05
  Dating                           -.09
  Fertility Issues                  .01
  Behaviors and                     .17
    Feelings
  Dangers                          -.11
  Condoms                          -.18
  Contraception                    -.09
  Abstinence                        .11
  Quality                          -.16 *
    Step 1 [R.sup.2]                .00
    Step 2 [R.sup.2]                .11 **
    [DELTA][R.sup.2] (Step 2)       .11 ***

Note. Sample size ranges from 170 to 178 because of missing data.

* p < .07. ** p < .05. *** p < .01. **** p <.001.